energy cost pass-through and strategic pricing: sectoral ...€¦ · by industry through a...
TRANSCRIPT
1
Energy Cost Pass-Through and Strategic Pricing:
Sectoral Evidence for the EU ETS
Victoria Alexeeva-Talebi
Centre for European Economic Research (ZEW), Mannheim, Germany
E-mails: [email protected]
Preliminary version
Abstract.
Price adjustments, particularly the pass-through relationships, are at the core of the analysis on how asymmetric climate change policy initiates two channels of carbon leakage in the energy-intensive sectors: (decreasing) market shares and profit margins. Under conditions of oligopolistic competition with strategic interactions, firms are in position to charge a flexible mark up over marginal costs to balance the negative impacts on both dimensions. Our empirical results demonstrate that strategic interactions of German producers with foreign competitors do matter in the EU ETS sectors. While energy cost shocks are largely born by the producers in the majority of the sectors, different patterns of asymmetric adjustment to the competitor’s prices might increase or decrease profit losses induced by the incomplete pass-through rate. The policy implications of the results are that strategic interaction between domestic and foreign firms could be a critical factor in applying offsetting instruments to address carbon leakage domestically. Consequently, accounting for oligopolistic structures (with and without strategic interactions) should be a central issue within the broader context of how market structure affects the climate change policies.
JEL Classification: F18, C22, L11
Keywords: Energy Cost Pass-Through, Strategic Oligopoly, Emissions Trading Scheme
Acknowledgment: The author is grateful to Ulrich Oberndorfer for valuable scientific advice. Funding by the European Commission under the grant agreement ENT/CIP/08/C/N02S00/1 is gratefully acknowledged.
Centre for European Economic Research (ZEW), P.O. Box 103443, 68034 Mannheim, Germany. Phone: +49
621 1235 214. Fax: +49 621 1235 226. E-mail: [email protected]
2
1. Introduction
In December 2008, the European Council decided upon the Energy and Climate Package – a
comprehensive legislative initiative proposed by the European Commission to achieve the
EU’s target of emissions reductions by at least 20% by 2020 (EU, 2008, 2008a). Recognizing
that asymmetric climate actions might lead to competitiveness distortions and carbon leakage,
new legislation proposal foresees that installations from energy-intensive and export-oriented
sectors receive preferential treatment in the third phase of the EU ETS from 2013 on. The
main options to address competitiveness-driven carbon leakage include free allocation of
allowances to existing and new facilities, financial compensation, border tax adjustments
(BTAs) and global sectoral agreements, i.e. instruments encouraging sector-based activities in
developing countries. After all, the European Council favoured free allocation to address
competitiveness loss and carbon leakage issues in the European Union. But global sectoral
approaches – as alternative option to reduce carbon leakage – might still apply to European
facilities as Part of a Post 2012 Framework.
For policy-makers, the possibility of sector-specific carbon leakage is at the centre of
discussions on how to effectively reduce emissions under domestic and international climate
policies. Sound understanding of its drivers is indispensable for design of appropriate
countermeasures. In the context of international climate policy, the issues of
competitiveness-driven leakage have therefore triggered comprehensive research work. The
CGE-related literature mainly focused on assessing competitiveness and leakage effects
associated with the implementation of the EU ETS in the two first trading periods up to 2012
(Bollen et al., 2003; Klepper and Peterson, 2004; COWI, 2004; Reinaud, 2005; Peterson,
2006). Assuming that an increase in marginal carbon cost is fully borne by consumers1 of the
final good, these studies estimate how domestic suppliers adjust market shares on both
domestic and foreign markets.2 However, the cost pass-through literature suggests that cost
increases under uneven carbon constraints would not necessary be fully passed onto the
consumers of energy-intensive goods through product price increases but are rather absorbed
by industry through a reduction of profit margin – in the extreme case, this might imply
constant prices and sustaining output level but decreasing profit margins (Reinaud, 2005a and
2005b, Smale et al. 2006, McKinsey and Ecofys 2006, Hourcade et al. 2007, Öko-Institut
1 In this case the profit margin of the producers remains unchanged.
2 Reinaud (2008) refers to changes in international trade flows of carbon constrained products as a short-run
leakage route (channel).
3
2008, CE Delft 2008). Differences in returns on capital between constrained and
unconstrained producers give room for leakage by creating incentives to relocate the business
abroad. Studies using detailed global sectoral models (e.g. Quirion 2003, Demailly and
Quirion 2006a, Hidalgo, Szabo et al. 2005, Palmer et al. 2006) addressed sectoral carbon
leakage referring explicitly to cost pass-behaviour of selected energy-intensive sectors.
Assuming a range of cost-pass through rates (and other parameters such as trade elasticity),
these studies estimated the impact of carbon price on both market share and profits of
European energy-intensive industries. Given the fact that results crucially depend on the
ability to pass-through additional costs, empirical foundation of such assumptions is
fundamental for any conclusions on appropriate countermeasures. Surprisingly, ex-post
studies providing evidence on the cost pass-through behaviour of European energy-intensive
sectors other than power are at best scarce (see for example Walker et al. 2006 on cement).
The purpose of this paper is to analyse the potential passing through capacity of the additional
energy cost in German energy-intensive industries covered by the European emission trading
scheme (EU ETS). Within our stylized theoretical and empirical framework, we employ a
variant of the mark-up model of price determination which allows for possible strategic
interaction between domestic and foreign firms in the form of limiting the impact of cost
shocks on price competitiveness. The key element of the model is that firms would be in
position to charge a flexible mark up over marginal energy costs to earn profits and to protect
market shares. In this setting, firm’s decision on how to adjust market shares and profit
margins is endogenous to a particular policy shock. Although strategic interactions in energy-
intensive sectors might be very relevant, the empirical cost-pass through literature does not
typically take such effects into consideration. The few studies accounting for the competitors'
prices have been limited to the analysis of exchange rate pass-through for exporters in
selected manufacturing industries (e.g. in automobile sector in Switzerland: Gross and
Schmitt 2000, in ceramic tiles industry in Italy and Spain: Balaguer etr al. 2003). To our
knowledge, there is no empirical study which analyses the capacity of German energy-
intensive sectors to pass-through costs accounting for possible oligopolistic interactions with
foreign competitors. The model is estimated for 17 energy-intensive sectors at the 3-digit-
level – the corresponding sectors at the lower level of sectoral disaggregation are chemicals,
paper, metals, refineries and non-metallic products. The focus on German energy-intensive
sectors is to account for these industries being the biggest producer within the EU-27
(Eurostat, 2007).
4
The structure of the paper is as follows: Section 2 sets out the theoretical framework for
measuring the incidence of energy price shocks in the presence of oligopolistic structures with
strategic interactions. Section 3 discusses the econometric procedure and the data base.
Section 4 presents and analyses the results. Section 5 concludes.
2. Model
To analyse the potential passing through capacity of additional energy costs in German
energy-intensive sectors, we employ a variant of the mark-up model of price determination
built upon the work of Dixit-Stiglitz (1977) and Dornbusch (1987)3. Under conditions of
imperfect competition in heterogeneous goods, this framework allows for possible strategic
interaction between domestic and foreign firms. The key element of the model is that firms
would be in position to charge a flexible mark up over marginal energy costs to earn profits
and to protect market shares even. The main purpose of this section is to give a theoretical
underpinning to the empirical model estimated subsequently.
Assume that the representative consumer maximizes the following separable utility function:
[ , ( )]u U z V x (1)
where z and x are two commodities and x is an index of different brands of the same
commodity. Suppose that there are Dn identical domestic firms and Fn foreign firms in (our)
home market – the latter are identical to each other but not to the domestic firms.4 The
demand for each individual variant is obtained by maximizing the utility function:
c
d dx x P p ; 1
(1 )c
(2)
The price index of the variants of x is given as:
1
1 1
D F hn nh hd f
d f
P p p
;
(1 )h
(3)
where ip denotes the price of a domestically produced variant and jp denotes the price of an
imported variant.
3 Dornbusch (1987) considers the Dixit-Stiglitz model (1977) to capture the effects of imperfect competition and
product differentiation on the output price responses to exchange rate changes.
5
Under conditions of imperfect competition, assume too that individual firms are sufficiently
large to affect the industry price P , while strategic interactions between the firms are
introduced by way of a conjectural variation ( 0 1 ). The latter parameter indicates
that firms respond to a one-percentage-point rise in the industry price by increasing their
prices by percent5. The profits of the Dn domestic firms are given by:
d d d dp e x (4)
where ix are the outputs, and ie are the unit energy costs of the domestic firm.
The first-order condition of profit maximization for an individual domestic producer k
becomes:
0k kk k k
k k
x x Px p e p P p
(5)
Thus, a single firm’s production volume is affected directly via change in it’s individual price
k kx p and indirectly via changes in the output price index resulting from his own decision
k kx P P p .
Let now denote the elasticity of the aggregate price level with respect to the single
supplier’s own price:
k kdP P dp p (6)
Since the individual firm has to take into consideration the extent to which his action affects
the output price index P , this term captures the strategic interaction between firms as
perceived from the domestic firm k . Using the above definition for , the first-order
condition can be simplified to:
1 ( ) ( 1) 0k k kp e c p (7)
and solved for the optimal price under strategic interaction:
4 Assume further that there is an effective separation between the home and foreign markets. Thus, it’s possible
to separately discuss the pricing behaviour of foreign producers in our market.
5 In the Cournot model of imperfect competition in homogenous goods (perfect substitutability between the
domestic and imported goods), a firm’s mark-up depends on it’s market share. Firms with a high market share
are considered to be able to charge higher prices. But in reality, this might be difficult if competitors are not
expected to follow a firm’s price increase. Hence, firm’s optimal pricing strategy will not only depend on it’s
6
11
(1 )
kk k
ep e
c
(8)
The mark-up pricing equation (8) highlights the fact that firm’s optimal price policy is no
longer to charge a constant but rather a flexible mark-up over energy costs. Assuming that
conjectural variation for all firms i and j other than k is given by:
, ,d k f d k fdp p dP P (9)
one gets the following expression for the elasticity 6:
1
(1 )[ ( ) ]D F hd fn n p p
(10)
The individual domestic firm’s price reaction is then a function of relative prices, the
conjectural variation and the elasticity of substitution among variants.
( , , )d d f dp F p p c e (11)
By following similar steps (as in the case of the domestic firm) one gets the foreign firm’s
reaction function:
( , , )f f d fp F p p c e (12)
From equations (11) and (12), it’s clear that optimal pricing decisions of domestic and foreign
firms are interdependent.
In the context of the unilateral EU climate change policy, this simple framework allows
illustrating important insights for the aggregate price level. Figure 1 and Figure 2 demonstrate
these implications for two alternative policy options being recently under consideration in the
EU. The curves AA and A*A* are the price reaction functions of a representative domestic
and foreign firm, respectively. B is the initial equilibrium with carbon costs being reflected in
the energy prices of the domestic firm only. Now consider the case (Figure 1) where home
country intends integrating foreign producers into the domestic emission trading scheme or
imposing the import tariff on the foreign products in the domestic market (see for further
details Alexeeva-Talebi et al. 2008). This policy option will shift the foreign reaction function
market share but be conditioned by the anticipation of competitor’s reaction to this strategy. This interrelation is
expressed as conjectural variation.
6 In the Dixit-Stiglitz model (1977) this elasticity is zero.
7
up and to the right while leaving the domestic reaction function unchanged. The new
equilibrium B’ is characterized through an increase of the aggregate price level while the
domestic firm responds with a higher price and mark-up to this policy chock. Alternatively,
home government intends to subsidize a fraction of the carbon costs getting reflected in the
lower domestic energy prices (Figure 2). Such a policy option will shift the domestic reaction
function down and right while leaving the foreign country’s reaction function in place. The
new equilibrium is therefore at B’’ with a lower aggregate price level.
This framework allows us to test if German energy-intensive sectors price strategically –
through variation in their mark-ups – in the face of the energy cost shocks. Taking logarithms,
we receive the solution out of equations (8) and (10) for the domestic supplier:
(1 )d fit i it i itp e p ,
1i
ii
(13)
If i is zero, domestic prices are set exclusively with respect to the domestic producer’s cost
situation. This would be exactly the case if the Dixit-Stiglitz model (1977) holds, reflecting
the constant mark-up on domestic marginal energy costs. If i is one, the domestic prices are
set exclusively with respect to the foreign producer’s prices in the domestic market. If i
varies between zero and one, then domestic prices react to both domestic unit costs and
foreign competitor’s prices. The larger the competitive pressure i is, the higher
substitutability between domestic products and foreign export goods, the larger i is. In this
particular case the domestic producers gear to a much higher extend to the competitor’s prices
B
A*
A’
A
o
fp '
fp
fp
o
dp
'
dp
B’’ B
A* A*’
A
A* A*’
o
fp '
fp
fp
o
dp
'
dp
B’
dp
Figure 1: Increasing aggregate price Figure 2: Decreasing aggregate price
8
than to their own cost changes. The aim of the subsequent section is to determine the
behavioural parameter of the in the domestic empirically.
3. Data and econometric procedure
The equilibrium price equation (13) is estimated for German energy-intensive sectors covered
by the emissions trading scheme (EU ETS) with monthly data for the period January 2000 to
December 2008. The analysis is carried out for 17 sub-categories at the 3-digit-level
according to the German Product Classification for Production Statistics 2002 (GP 2002) –
the corresponding sectors at the lower level of sectoral disaggregation are basic chemicals,
rubber and plastic, paper and pulp, publishing, basic metals, refineries and non-metallic
products. The selection of the respective energy-intensive sub-sectors is based on Graichen et
al. (2008) – this publication specifies industrial branches participating in the emission trading
scheme (EU ETS) in Germany7 (see Table 1).
Time series for the sectoral domestic and foreign competitor’s prices stem from German
Statistical Office (Statistisches Bundesamt, 2009a). The former is domestic output price index
(producer price index) for goods in respective sub-sectors sold on the German market. The
latter measures the price development of the imported goods purchased in Germany from
non-domestic producers.8 Since more frequent price data (e.g. on the weakly base) are not
available for these sectors and the EU ETS is still in an early stage (i.e. respective time series
on carbon prices is relatively short), the passing-through capacity of additional carbon prices
cannot be directly estimated. However, the carbon costs are likely to be incorporated in the
electricity prices (Zachmann and Hirschhausen, 2008), while carbon regulations might have
significant feedback effects on the international price development of the primary energy
carriers coal, gas and oil (Böhringer and Vogt, 2003). Therefore, we analyse the potential
passing through capacity of additional energy costs in German energy-intensive sectors, using
energy price indices for electricity, natural gas, hard coal and crude mineral oil from German
Statistical Office as proxies (Statistisches Bundesamt, 2009b). Due to the data availability, the
prices for industrial customers are used for electricity only. The data for crude oil, natural gas
7 Graichen et al. (2008) list industrial braches at the 4-digit level in NACE classification. Due to the data
availability, the analysis has to be carried out at a lower level of disaggregation. Therefore, we assign industrial
braches at the 4-digit level in NACE classification to the 3-digit level in GP 2002 classification. Due to the data
lack, we do not, however, consider the sectors manufacture of coke oven products (231) and manufacture of
bricks, tiles and construction products (264).
8 The appropriate price is the C.I.F. price (cost, insurance, freight) at the border.
9
and hard coal are import indices. All data series are (seasonally unadjusted) indexes with the
2000 monthly average as the base value.
Table: Industrial braches participating in the EU ETS in Germany
Sector Code GP 2002
Manufacture of pulp, paper and paper products 21
Manufacture of pulp, paper and paperboard 211
Manufacture of articles of paper and paperboard 212
Publishing, printing, reproduction of recorded media 22
Printing and service activities related to printing 222
Manufacture of coke, refined petroleum products and nuclear fuel 23
Manufacture of refined petroleum products 232
Manufacture of chemicals and chemical products 24
Manufacture of basic chemicals 241
Manufacture of pharmaceuticals, medicinal chemicals and botanical products 244
Manufacture of soap, detergents, cleaning, polishing 245
Manufacture of other chemical products 246
Manufacture of rubber and plastic products 25
Manufacture of rubber products 251
Manufacture of plastic products 252
Manufacture of non-metallic mineral products 26
Manufacture of glass and glass products 261
Manufacture of non-refractory ceramic products 262
Manufacture of ceramic tiles and flags 263
Manufacture of cement, lime and plaster 265
Manufacture of other non-metallic mineral products 268
Manufacture of basic metals 27
Manufacture of basic iron and steel and of ferrous-alloys 271
Manufacture of basic precious and non-ferrous metals 274
We start our econometric procedure by testing the properties of the time series. The results
from the Dickey-Fuller unit root test suggest that none of the dependent and explanatory
variables are stationary in levels. Achieving the stationarity after differencing once implies
that they are non-stationary processes of order one (I(1)).9 Subsequently, we tested for the
existence of a long-run relationship using the Johansen maximum likelihood procedure. The
existence of cointegration relationship between domestic and foreign prices has been rejected
for few product categories, making the estimation in first differences necessary for all time
9 The results are available upon the request.
10
series. This might be due to the fact that the data set is relatively small so that the probability
to identify the long-run properties of the data is reduced. Under this approach, the long-run
effects of the determinants are captured by summing up the coefficient of the current and
lagged first differences. Thus, the choice of the numbers of lags included in the regression
might be crucial for the results. According to the Schwartz’ Bayesian Information Criteria
(BIC), the appropriate number of lags is one for all time series. Akaike’s Information
Criterion (AIC) suggests specifying the model with a one lag for each explanatory variable for
the most but not for all time series. In few sectors, the latter suggests specifying the model
with a higher number of lags (up to six). Obviously, the firms may delay adjustments of their
producer prices to the cost shocks. To keep the degree of freedom as high as possible, the
number of the lags has been restricted to four. Model specifications including one and four
lags are reported in the Appendix. The model specification with a high number of lags is used
to check the model robustness.
The autoregressive distributed lag representation of equation (13) has been estimated for 17
sub-sectors using the energy prices for electricity, crude oil, natural gas and hard coal. In
order to asses the vulnerability of domestic energy-intensive sectors to the foreign
competitors, we account for possible asymmetric reactions of German producers to the
foreign competitor’s price separating the months with positive and negative price changes:
, 0
0,
f ff t t
t
p if pp
otherwise
, 0
0,
f ff t t
t
p if pp
otherwise
and estimate the following equation in first differences including up to the fourth lag:
, , , , , , , , ,1 1 1 0
'n n n n
d d f fi t i i l i t l i l i t l i l i t l i y t y i t
l l l y
p p p p e
(14)
with ' ( , , , )o c e gt t t t te p p p p ,
where i is the constant, ,i t is the error term and 'te is the vector of energy prices
with , , ,o c e gt t t tp p p p being the price for oil, coal, electricity and gas, respectively.
11
4. Results
The results from the model specification with one lag (Tables 2 and 3) and four lags (Tables 4
and 5) together with diagnostic statistics are presented in the Appendix. To check the
robustness of our results, we additionally estimated for every sub-sector a model with a higher
number of lags (up to six) – those results will be reported only if additional insights have been
obtained.10 To emphasise the heterogeneity in the price-setting behaviour – and therefore the
value of analysing the cost pass-through relationships at the highest possible level of
disaggregation – we report the results for all sub-sectors.
Paper and pulp (21) and printing (22)
Our results on the ability to pass-through additional energy costs and strategic pricing are
rather heterogeneous in both goods categories.
For the manufacturers of articles of pulp and paper (212), Tables 2 and 4 indicate that
passing-through additional energy costs is being completed within few months: In the model
specification including four lags, the respective coefficients – with one exemption – have
expected sign and are significant on the 5% significance level. Given the fact that coal is not
used in the production process (Table 1), any price increases of this input factor are not
reflected in the output prices. Empirical evidence has been found for the existence of the
flexible mark-up as domestic producers decrease own prices in the event the competitor’s
increase them. In the model specification with a higher number of lags, the weak evidence (at
the 10% significance level) for adjustments to the negative price shocks of the competitors
has been also received. Thus, this sub-sector lets profits bear the brunt of adjustments to the
competitor’s price changes as increasing market share comes at the expense of the decreasing
profit margin.
In contrast, the ability of the producers of paper and pulp (211) to pass-through cost shocks is
more limited: According to our estimates in the specification including one and four lags,
output prices of these products rise due to the increased gas prices. Data on the energy
intensity with respect to the input factor gas suggest, however, that pass-through rate of these
cost shocks is complete (Table 1). Evidence for complete passing-through oil price increases
has been found in model specification with a higher number of lags, but price shocks of the
most important energy input factor electricity are born by the producers even within a time
horizon of six months. The domestic firms appear to partly offset these losses in profit margin
10 Results in the model specification with a higher number of lags (up to six lags) are available upon request.
12
via increasing prices in the event the competitor’s increase own prices: Following the 1%-
positive competitor’s price shock, manufacturers of paper and pulp increase the domestic
prices immediately (within one month) by up to 0.47%.
For the printing sector (222), we do not find the evidence that any types of energy cost shocks
might be passes-through to the consumer in the short-run (within four months). According to
the model specification with a higher number of lags (up to six lags), this sector is capable to
more than completely pass-through gas price shocks, while price increases of two other
energy input factors (oil and electricity) are born by the producers. The latter model
specification provides evidence on strategic pricing of the firms, i.e. there are price
adjustments to both competitor’s price increases and decreases. Obviously, such price-setting
strategy might be suitable to maintain the market share in the domestic market. Adjusting to
increasing and decreasing competitor’s prices will, however, differently affect the profit
margin of the respective firms, albeit the scope to compensate the incomplete pass-though rate
appears to be limited in this sub-sector. Finally, own price adjustment appears also to be an
important driver of the price-setting strategy in this sub-sector.
Basis chemicals products (24)
None of the sub-sectors producing chemical products are capable to completely pass-through
all types of relevant energy prices shocks within the time horizon of six months. The speed
and scope of adjustments differs across those sectors. However, the most of the sub-sectors
are capable to pass-through a substantial fraction of additional energy price increases over this
time horizon.
Manufactures of basic chemicals (241) rapidly – within one month – pass-through a
considerable part of energy price shocks to the consumers. The corresponding coefficients are
highly significant for electricity and oil price increases and weak significant for coal price
increases. We have not found any evidence that gas price shocks – with gas being one of the
major energy inputs in the production – cannot be passed-through over this time horizon
(Table 1). The model specification with one lag provides also strong evidence on the strategic
interactions with foreign competitors: Domestic producers let the profit margin bear the brunt
of adjustments to the competitor’s prices decreasing the prices in the event the competitors let
them fall. The model specification with a higher number of lags does not significantly change
our results.
The manufacturers of soap & cleaning (245) and other chemical products (246) are capable to
pass-through a substantial fraction of additional energy costs too but over a longer time
13
horizon: In the model specification with one lag, the respective coefficients in the former sub-
sector are insignificant. In the model specification with a higher number of lags (up to six),
producers of soap & cleaning products are able to pass-through the biggest fraction of energy
price shocks – i.e. increasing electricity and gas prices – to the consumers. Oil price increases
are thereby born by the producers. Strategic pricing in this sub-sector enclose maintaining the
market share via adjustments to both positive and negative price movements of the
competitors. For the latter sub-sector, we have found robust empirical evidence – i.e. in the
model specifications with a different number of lags – for potential to pass-through additional
gas prices. But our results regarding the ability to pass-through additional electricity prices are
less clear and depend on the number of lags included in the model specification. According to
the Table 1, only these both types of energy are being used in the production process. Thus,
the ability of the producers in this sub-sector to pass-though the energy price shocks varies
from being substantial and to being complete. The results regarding the existence of the
flexible mark-up in this sub-sector are ambiguous too as respective coefficients differ in
model specifications with an alternative number of lags.
In contrast, the ability to pass-through energy cost shocks in the production of
pharmaceuticals (244) remains rather limited even over a time horizon of six months. This
sub-sector is capable to completely pass-through the gas price increases only, while other
price increases – i.e. electricity, oil and coal – are born by the producers. In the model with a
higher number of lags, our results indicate that producers of pharmaceuticals follow negative
price shocks of the competitors. Obviously, such strategic price adjustments put an additional
burden on the profit margin in this sub-sector.
Rubber and plastic products (25)
Within this sub-sector, we find even a more pronounces heterogeneity: Producers of rubber
(251) are able to more than completely pass-through all relevant types of energy price shocks
in the short-run, within four months (Tables 2 and 4). Adjustments to the positive price
shocks of the competitor’s are weak significant in the model specification with four lags (at
the 10% significance level) and significant in the model specification with a higher number of
lags (at the 5% significance level), indicating that producers might be able to get additional
profits at the expenses of decreasing market shares. In contrast, producers of plastic (252) are
much more limited in their ability to pass-through energy price shocks. In the model
specification with a higher number of lags, the biggest fraction of the energy price shocks is
born by the producers even 6 months after the price shocks – the electricity and gas price
14
shocks cannot be passed-through. We find a weak evidence for passing-though oil price
increases. In the model specification with 6 lags, the adjustments to the decreasing
competitor’s prices (in the fifth month after the price shock) alleviate the negative
implications for the profit margin.
Non-metallic mineral products (26)
Our results in Tables 3 and 5 suggest that only one sub-sector – i.e. manufactories of non-
refractory ceramic products (262) – completely passes-through all energy costs increases in
the short-run, while other sectors are capable to pass-through at least a (significant) fraction of
the energy price shocks.
Three out of five sub-sectors in this product category – manufactures of non-refractory
ceramics (262), ceramic tiles & flags (263) and cement & lime (265) – are able to pass-
through electricity price shocks to the domestic consumers within few months, albeit the
results for the producers of ceramic tiles and flags are significant in the model specification
with one lag only. The corresponding coefficients are significant on the 5% significance level,
with the highest value of up to 0.55% in the sub-sector manufacture of cement & lime. Table
1 suggests further that those producers are able to pass-through more than 100% of the
electricity price increases. Surprisingly, we do not find – with an exception in the model
specification with one lag for the producers of non-refractory ceramic goods and a model
specification with a higher number of lags (up to six) for the glass producers (261) – any
evidence for passing-through gas price shocks, although gas constitutes one of the most
important energy production factors (Table 1). The (weak) evidence for passing-through oil
price changes has been found for manufactures of non-refractory ceramics and other non-
metallic products. Finally, two sub-sectors – producers of glass and cement & lime – pass-
through additionally coal price shocks, with coefficients ranging between 0.04% and 0.06%.
This is remarkable as – at least in 2002 – this production factor has not been used in both
sectors, while the change in the production function might deliver an explanation for this
result.
To sum-up, producers of non-refractory ceramic goods are capable to completely pass-though
all energy price shocks within four months. Producers of cement & lime pass-through the
biggest fraction of energy price increases in the short-run: The results are robust for the
passing-through electricity and – in the model specification with a higher number of lags – oil
price increases. We have also found evidence for passing-through gas and oil prices to the
consumers, albeit these results are not robust, i.e. they depend on the number of lags included
15
in the model specification. In contrast, producers of glass products and ceramic tiles & flags
bear the biggest fraction of energy price increases even within the time horizon of six months.
The results for producers of other non-metallic products (268) are ambiguous, depending on
alternative model specifications.
Our results in Table 3b indicate further that domestic price-setting of German non-metallic
mineral producers might be affected by the negative price movement of the competitors in
two sub-sectors: Those sectors are manufactures of cement, lime and plaster and glass
producers. Such price adjustment puts an additional pressure on the profits reducing the profit
margins in the respective sub-sectors. In the model specification with a higher number of lags
(up to six lags) the adjustment to the negative price movements has been found for the
manufactures of other non-metallic mineral products. Our hypothesis about the existence of
the strategic interactions within the extended Dixit-Stiglitz framework (i.e. existence of the
flexible mark-up) in Section 2 cannot be confirmed for the remaining sub-sector (263 and
262), as the respective coefficients (competitor’s prices) are not statistically significant. In the
standard version of the Dixit-Stiglitz model (1977) the oligopolistic firms are significant small
– they are not capable to make any impact on the aggregate (industry) price level and
therefore do no explicitly consider the prices of the competitors. Thus, one reason for
competitor’s prices being non-significant in this sector might be that printing firms are
significant small and do not possess market power.
Ferrous and non-ferrous metals (27)
According to the Tables 3 and 5, manufacturers of ferrous (271) and non-ferrous metals (274)
are capable to completely pass-through only a limited amount of energy price increases in the
short-run. Producers of ferrous metals pass-through coal price shocks – this appears to be
important as Table 1 suggests that coal represents the most significant energy input factor in
this sub-sector – and oil price increases. Given the energy intensities in the Table 1, the
biggest fraction of energy costs – gas and electricity – is being born by the producers even
within the time horizon of six month. Producers of non-ferrous metals are capable to pass-
through even a smaller fraction of additional energy prices, namely oil price shocks only.
Thus, the biggest fraction of price increases – particularly of the most important energy input
factor electricity – is born by the producers in this sub-sector. Further experimental runs –
model specification including up to six lags – do not significantly change our results
indicating that a significant fraction of energy price shocks is born by the producers of both
sub-sectors even within the time horizon of up to six months.
16
Further, Tables 3 and 5 provide strong evidence that price-setting behaviour of German
producers of ferrous and non-ferrous metals is largely determined by the evolution of
competitor’s prices. Thereby, these sectors adjust to both types of competitor’s price shocks,
i.e. to the positive and negative price movements. For both sectors, our estimation results
report positive asymmetric short-run pricing behaviour with respect to the competitor’s
prices: Positive competitor’s price shocks have a stronger positive influence on prices of
producers of ferrous metals, than negative shocks (i.e. competitor’s prices decrease). The
estimated (lagged) coefficient for positive competitor’s price changes is larger than those for
negative ones ( 1 1ˆ ˆt t ), while the values amount 0.98% and 0.56%, respectively. Thereby,
prices are adjusted almost immediately, i.e. within one month (Table 5). For producers of
non-ferrous metals, we have found even a more pronounced positive asymmetric pricing.
Following the 1%-positive competitor’s price shock, manufacturers of non-ferrous metals
increase the domestic prices by 1.14%. Domestic prices of these products decrease by 0.92%
only, if competitor’s prices drop by 1% (Table 5). Obviously, such price-setting strategy
might be suitable to maintain the market share in the domestic market. Adjusting to increasing
and decreasing competitor’s prices will, however, differently affect the profit margin of the
respective firms. Adjusting to increasing prices of the competitor’s prices allows German
producers to partly offset the profit losses induced by the incomplete pass-through rate.
Manufactures of refined petroleum products (23)
According to our results, the producers can pass-through the 1% increase of the oil prices by
0.39% to the consumers immediately within one month. In the model specification with four
lags, the manufactures are capable to pass-through up to 0.71% of the oil price increase. The
latter model specification provides also a weak evidence for passing-through electricity and
coal price increases. Given the fact that Table 1 does not provide the energy intensities for
this sub-sector for the confidentiality reasons, the energy cost pass-through rate cannot be
estimated. Finally, the results on strategic pricing are not robust with respect to the model
specification.
17
5. Conclusions
This paper analyses the potential passing-through capacity of the additional energy cost
shocks in German energy-intensive industries covered by the EU ETS within a theoretical and
empirical framework. It appears to be important to account for the price-setting behaviour of
the producers at the lowest possible level of sectoral disaggregated. Our results indicate that
sub-sectors might be rather heterogeneous in price adjustments (including cost-pass-through
relationships). Our main findings might be summarized in the following way:
First, manufacturers of three (out of seventeen) product sub-categories – that are
articles of paper and pulp, rubber and non-refractory ceramics – are capable to
completely pass-through all types of relevant energy price increases in the short-run.
In nine sub-sectors – that are paper and pulp, refined petroleum, basic chemicals, other
chemical products, soap & cleaning, glass, ceramic & tiles, other non-metallic mineral
products, iron & steel – producers are capable to pass-through a significant fraction of
energy price within this time horizon. In all remaining energy-intensive sectors, the
potential to pass-through electricity, coal, gas and oil price increases is rather limited,
albeit every sector is capable to completely pass-through one type of energy price
shocks in the short-run. Thus, those four sub-sectors – that are manufacturers of
printing service, pharmaceuticals, plastic and non-ferrous metals – bear the biggest
fraction of energy price increases even within the time horizon of six months.
Competitive pressure from the foreign competitors and the existence of long-term
contracts with suppliers might provide the explanation for these results. The latter
appears to be plausible as none of these sub-sectors passes-through the price increases
of the most important energy input factors in the respective production process.
Second, the most of the German EU ETS sectors appear to have a flexible mark-up
over marginal (energy) costs – thus, strategic interactions with foreign competitors do
matter to balance negative implications for profit margin and market shares. Strong
evidence for existence of a flexible mark-up has been found for the producers of basic
metals, paper and pulp, printing services and basic chemicals. Strong evidence is
given if respective coefficients are statistically significant at the 5% level and results
are robust in the model specifications with different number of lags. Weak evidence
for flexible mark-up exists for producers of articles of paper, refined petroleum,
pharmaceuticals, soap & cleaning, other chemical products, rubber, plastic, glass,
cement and other non-metallic products. In our setting, there is weak evidence if
18
coefficients are statistically significant at the 10% level and/or sensitive to the number
of included lags within the model specification. Manufacturers of non-refractory
ceramics and ceramic tiles & flags appear to have a fixed mark-up as they do not take
into consideration the prices of the foreign competitors. However, our theoretical
framework suggests that strategic interactions between domestic and foreign
producers under conditions of imperfect competition will limit the ability of a
producer to pass-through the energy cost shocks. Energy cost shocks cannot be
completely passed-through in (almost) all sectors (see above), while sectors where
strategic interactions exist do not perform significantly different in this respect as
sectors without any strategic interactions. One of the explanations for this result might
be a different time horizon: While our data allows performing a short-run analysis
only, the theoretical model is a long-run framework.
Third, industries with strategic interactions demonstrate different patterns of
asymmetric adjustments to the competitor’s prices: Adjusting to negative competitor’s
price changes might additionally decrease profit losses induced by the incomplete
pass-through rate, while adjusting to positive competitor’s price changes might
decrease profit losses. The price reactions towards the competitor’s prices might be
therefore used as an additional criterion for the vulnerability of the EU ETS sectors.
Our main insight is that hypothesis of strategic interactions under oligopolistic competition
holds for the most of the EU ETS sectors in Germany. In this respect, our results are very
much related to the findings of Quirion (2008) who conducted a simulation analysis for iron
and steel producers. Assuming oligopolistic competition (without strategic interactions and
thus with constant mark-ups), the study assesses the implications of the EU ETS on the
production level (market share). Under conditions of oligopolistic competition with strategic
interactions, however, firms are in position to charge a flexible mark up over marginal costs
to balance the negative impacts on market share and profitability. In this setting, firm’s
decision on how to adjust market shares and profit margins is endogenous to a particular
shock (see again section 2). The policy implications of the results are that strategic interaction
between domestic and foreign firms could be a critical factor in applying offsetting
instruments to improve competitiveness of domestic energy-intensive industries.
Consequently, accounting for oligopolistic structures (with and without strategic interactions)
should be a central issue within the broader context of how market structure affects the
climate change policies. Particularly, our results suggest that sector-specific choice of the
instruments to address the leakage problem might be a promising approach.
19
The further research might therefore focus on the following issues: (i) better understanding of
implications of the existence of flexible mark-up for alternative offsetting measures –
simulation models might provide an appropriate methodological framework, (ii) cost-pass
though issues in the long-term perspective and (iii) better understanding why particular price
shocks (types of energy) are passed-through in some sectors in the short-run but not in the
others.
6. References
Anger N, Böhringer C, Moslener U (2007) Macroeconomic Impacts of the Clean
Development Mechanism: The Role of Investment Barriers and Regulations. ZEW
Discussion Paper No. 07-026, Mannheim
Böhringer C. and C. Vogt (2003), “Economic and environmental impacts of the Kyoto
Protocol”, Canadian Journal of Economics, 36 (2), pp. 475–496.
Barzel, Yoram (1976), "An Alternative Approach to the Analysis of Taxation," Journal of
Political Economy, 84(6), pp. 1177–1197.
Dornbusch, R (1987). Exchange rates and prices, The American Economic Review 77 (1), 93-
106.
Johnson, Terry R., "Additional Evidence on the Effect of Alternative Taxes on Cigarette
Prices," Journal of Political Economy, December 1978, 86(2), pp. 325-328.
Knetter, M M (1994). Is export price adjustment asymmetric? Evaluating the market share
and marketing bottlenecks hypotheses. Journal of International Money and Finance 13
(1), 55-70.
Mahdavi, S (2000). Do German, Japanese, and U.S. export prices asymmetrically respond to
exchange rate changes? Evidence from aggregate data. Contemporary Economic
Policy 18 (1), 70-81.
Sumner Daniel A., "Measurement of Monopoly Behavior: An Application to the Cigarette
Industry," Journal of Political Economy, October 1981, 89(5), pp. 1010-1019.
Sumner, Michael T. and Robert Ward, "Tax Changes and Cigarette Prices," Journal of
Political Economy, 1981, 89(6), pp. 1261-1265.
20
Statistisches Bundesamt (2005), Produzierendes Gewerbe – Material- and Wareneingang im
Verarbeitenden Gewerbe sowie im Bergbau und in der Gewinnung von Steinen und
Erden 2002, Fachserie 4 / Reihe 4.2.4, Wiesbaden.
Statistisches Bundesamt (2009a), Preise – Index der Erzeugerpreise gewerblicher Produkte
(Inlandsabsatz) nach dem Güterverzeichnis für Produktionsstatistiken, Ausgabe 2002
(GP 2002), Lange Reihen von Januar 1995 bis Dezember 2008, Wiesbaden.
Statistisches Bundesamt (2009b), Data on energy price trends, Long-time series from January
2000 to December 2008, Wiesbaden.
Stiglitz, Joseph E., "Who Really Pays the Tax: Tax Incidence," in Economics of the Public
Sector, Chapter 17, 1988. New York, NY: W.W. Norton & Company, pp.411-436.
Sullivan, Daniel, “Testing Hypotheses about Firm Behavior in the Cigarette Industry,”
Journal 20 of Political Economy, 1985, 93(3), pp. 586-598.
Sung, Hai-Yen, The-Wei Hu, and Theodore E. Keeler, "Cigarette Taxation and Demand: An
Empirical Model," Contemporary Economic Policy, July 1994, Vol. 12, pp. 91-100.
Zachmann, G. and C. v. Hirschhausen (2008), First Evidence of Asymmetric Cost Pass-
through of EU Emissions Allowances: Examining Wholesale Electricity Prices in
Germany, Economics Letters 99, 465-469.
21
Table 1: Energy intensity in German EU ETS sectors (% of total production costs in 2002)
11 Total includes energy and other energy-related inputs (i.e. district heating). Source: Statistisches Bundesamt (2005) .
Code GP 2002 Sector Coal Gas Oil Electricity Total11
21 Manufacture of pulp, paper and paper products
211 Manufacture of pulp, paper and paperboard 0.8 5.1 0.4 5.7 13.0
212 Manufacture of articles of paper and paperboard 0.0 1.5 0.4 1.8 4.3
22 Publishing, printing. reproduction of recorded media
222 Printing and service activities related to printing 0.0 0.8 0.4 2.4 3.6
23 Manufacture of coke, refined petroleum products and nuclear fuel
232 Manufacture of refined petroleum products n/a n/a n/a n/a n/a
24 Manufacture of chemicals and chemical products
241 Manufacture of basic chemicals 0.1 2.5 0.6 3.5 8.7
244 Manufacture of pharmaceuticals. medicinal chemicals and botanical products 0.1 0.7 0.4 0.8 2.3
245 Manufacture of soap, detergents, cleaning. polishing 0.0 0.8 0.3 1.2 2.6
246 Manufacture of other chemical products 0.0 0.9 0.0 1.4 3.2
25 Manufacture of rubber and plastic products
251 Manufacture of rubber products 0.0 0.8 0.0 1.7 3.1
252 Manufacture of plastic products 0.0 0.6 0.4 2.4 3.5
26 Manufacture of non-metallic mineral products
261 Manufacture of glass and glass products 0.0 5.2 1.5 4.5 11.3
262 Manufacture of non-refractory ceramic products 0.0 5.3 0.5 2.6 8.5
263 Manufacture of ceramic tiles and flags 0.0 15.1 0.0 5.8 21.4
265 Manufacture of cement. lime and plaster 0.0 4.5 5.8 17.4 43
268 Manufacture of other non-metallic mineral products 0.5 2.3 1.5 2.2 6.5
27 Manufacture of basic metals
271 Manufacture of basic iron and steel and of ferrous-alloys 7.5 4.4 0.9 5.2 18.3
274 Manufacture of basic precious and non-ferrous metals 0.1 1.2 0.2 3.2 5.1
22
Table 2: Regression results in the energy-intensive EU ETS sectors: GP code 21 - 24 (1 lag)
Dependent variable:
Sectors Paper and Pulp Printing Refineries Basic Chemicals
Code GP 2002 211 212 222 232 241 244 245 246
Constant -0.00** (0.00) 0.00** (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)
Own price-1 0.35*** (0.10) 0.37*** (0.10) -0.11 (0.10) -0.35*** (0.13) 0.05 (0.18) 0.09 (0.11) 0.03 (0.11) -0.21** (0.10)
Competitor’s price t=-1. positive 0.47*** (0.10) 0.02 (0.08) 0.07 (0.06) 0.24** (0.10) 0.27 (0.19) -0.04 (0.16) 0.02 (0.12) -0.03 (0.20)
Competitor’s price t=-1. negative 0.02 (0.11) 0.30 (0.23) 0.01 (0.05) 0.24** (0.10) 0.43** (0.17) -0.03 (0.19) -0.12 (0.22) 0.40 (0.27)
Electricity price t=0 0.07 (0.06) -0.01 (0.05) -0.03 (0.03) -0.27 (0.25) 0.00 (0.11) 0.04 (0.07) -0.06 (0.05) 0.14 * (0.08)
Electricity price t=-1 -0.01 (0.06) -0.04 (0.05) -0.02 (0.03) 0.42 (0.26) 0.28** (0.11) 0.01 (0.07) 0.08 (0.05) -0.02 (0.08)
Gas price t=0 0.05*** (0.02) 0.01 (0.02) 0.01 (0.01) 0.02 (0.09) 0.04 (0.04) 0.04 (0.02) 0.00 (0.02) 0.04 (0.03)
Gas price t=-1 -0.04 ** (0.01) 0.00 (0.02) 0.00 (0.01) -0.03 (0.09) 0.02 (0.04) -0.02 (0.02) 0.01 (0.02) 0.00 (0.03)
Oil price t=0 -0.01* (0.01) 0.00 (0.00) 0.00 (0.00) 0.39*** (0.02) 0.01 (0.01) 0.00 (0.01) 0.00 (0.00) -0.01 (0.01)
Oil price t=-1 0.00 (0.01) 0.00 (0.00) 0.00 (0.00) -0.02 (0.06) 0.04*** (0.01) -0.01 (0.01) 0.00 (0.00) 0.01 (0.01)
Coal price t=0 0.00 (0.01) -0.01 (0.01) -0.01 (0.01) -0.04 (0.06) 0.01 (0.03) -0.02 (0.02) 0.01 (0.01) 0.00 (0.02)
Coal price t=-1 -0.01 (0.01) -0.01 (0.01) 0.01 (0.01) -0.02 (0.06) 0.05* (0.03) 0.02 (0.02) 0.01 (0.01) -0.01 (0.02)
EU ETS dummy 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.02) 0.00 (0.00) 0.00*** (0.00)
Observation
R2
F-Test
AIC
BIC
106
0.58
10.59***
-8.13
-7.80
106
0.23
2.35***
-8.42
-8.09
106
0.10
0.92
-9.43
-9.11
106
0.81
33.01***
-5.02
-4.69
106
0.57
10.32***
-6.75
-6.43
106
0.09
0.79
-7.59
-7.26
106
0.13
1.11
-8.44
-8.11
106
0.24
2.46***
-7.36
-7.03
Note: *. ** and *** show significance at the 10%-. 5%-. and 1%-level. respectively. Standard errors are indicated in brackets.
23
Table 3: Regression results in the energy-intensive EU ETS sectors: GP code 25 -27 (1 lag)
Dependent variable:
Sectors Rubber and Plastic Non-metallic mineral products Basic Metals
Code GP 2002 251 252 261 262 263 265 268 271 274
Constant 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)
Own price-1 -0.05 (0.11) 0.35*** (0.12) 0.05 (0.10) 0.02 (0.11) -0.05 (0.10) 0.14 (0.12) -0.38*** (0.10) -0.18 (0.17) -0.66*** (0.17)
Competitor’s price t=-1. positive 0.06 (0.14) -0.09 (0.17) 0.24 (0.17) 0.03 (0.14) 0.14 (0.19) -0.03 (0.14) 0.23 (0.38) 0.90*** (0.21) 0.90*** (0.19)
Competitor’s price t=-1. negative 0.12 (0.17) 0.34 (0.22) 0.74* (0.42) 0.03 (0.20) 0.01 (0.28) 0.14** (0.06) 0.49 (0.73) 0.33 (0.23) 0.74*** (0.19)
Electricity price t=0 0.09* (0.05) -0.01 (0.04) 0.05 (0.06) 0.11* (0.05) 0.21*** (0.07) 0.38*** (0.10) 0.05 (0.14) 0.13 (0.22) -0.09 (0.28)
Electricity price t=-1 0.03 (0.05) 0.00 (0.04) -0.07 (0.07) -0.07 (0.06) -0.08 (0.07) -0.11 (0.12) -0.03 (0.15) -0.14 (0.23) 0.42 (0.29)
Oil price t=0 0.00 (0.02) 0.01 (0.01) -0.01 (0.02) -0.04 (0.02) -0.02 (0.02) -0.05 (0.04) 0.03 (0.05) 0.06* (0.08) 0.13*** (0.10)
Oil price t=-1 0.00 (0.02) 0.01 (0.01) 0.01 (0.02) 0.04 (0.02) 0.01 (0.02) 0.00 (0.04) 0.01*** (0.05) -0.04 (0.08) -0.09 (0.10)
Gas price t=0 0.00 (0.00) 0.00 (0.00) 0.00 (0.01) -0.01** (0.00) -0.01 (0.01) 0.00 (0.01) 0.04 (0.01) 0.04 (0.02) 0.09 (0.03)
Gas price t=-1 0.00 (0.00) 0.00 (0.00) -0.01 (0.01) 0.00* (0.01) 0.00 (0.01) 0.00 (0.01) 0.01 (0.01) 0.01 (0.02) -0.03 (0.03)
Coal price t=0 0.00 (0.01) 0.03*** (0.01) 0.05*** (0.02) 0.01 (0.01) 0.01 (0.02) 0.04* (0.02) -0.03 (0.03) 0.08 (0.05) 0.04 (0.07)
Coal price t=-1 0.02* (0.01) -0.01 (0.01) -0.04** (0.02) -0.01 (0.01) -0.03* (0.02) -0.01 (0.03) -0.02 (0.03) -0.07 (0.06) -0.01 (0.07)
EU ETS dummy 0.00 (0.00) 0.00 (0.00) 0.00* (0.00) 0.00 (0.00) 0.00** (0.00) 0.00** (0.00) 0.00*** (0.00) 0.00 (0.00) 0.00 (0.00)
Observation
R2
F-Test
AIC
BIC
106
0.11
1.00
-8.39
-8.06
106
0.26
2.75***
-8.92
-8.59
106
0.22
2.19**
-7.79
-7.46
106
0.10
0.90
-8.10
-7.77
106
0.22
2.16**
-7.66
-7.34
106
0.33
3.85***
-6.88
-6.55
106
0.26
2.71***
-6.23
-5.91
106
0.43
5.89***
-5.31
-4.99
106
0.42
5.62***
-4.81
-4.48
Note: *. ** and *** show significance at the 10%-. 5%-. and 1%-level. respectively. Standard errors are indicated in brackets.
24
Table 4: Regression results in the energy-intensive EU ETS sectors: GP code 21 -24 (4 lag)
Dependent variable:
Sectors Paper and Pulp Printing Refineries Basic Chemicals
Code GP 2002 211 212 222 232 241 244 245 246
Constant 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.01 (0.01) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)
Own price-1 0.11 (0.13) 0.19* (0.11) -0.16 (0.11) -0.21 (0.16) 0.17 (0.21) -0.05 (0.12) -0.10 (0.13) -0.15 (0.12)
Own price-2 0.18 (0.13) 0.18 (0.11) 0.15 (0.11) -0.30* (0.16) -0.12 (0.22) -0.07 (0.11) -0.07 (0.13) -0.20 (0.12)
Own price-3 0.10 (0.12) 0.16 (0.11) 0.24** (0.11) 0.06 (0.16) 0.15 (0.22) -0.11 (0.11) -0.14 (0.13) 0.00 (0.11)
Own price-4 0.12 (0.12) 0.14 (0.10) 0.15 (0.11) -0.11 (0.16) -0.36* (0.19) -0.06 (0.11) 0.00 (0.13) -0.09 (0.11)
Competitor’s price t=-1. positive 0.35** (0.15) 0.09 (0.11) 0.05 (0.07) 0.15 (0.12) 0.29 (0.22) -0.19 (0.19) -0.01 (0.16) -0.15 (0.23)
Competitor’s price t=-2. positive 0.11 (0.12) 0.14 (0.10) 0.04 (0.06) -0.03 (0.12) 0.00 (0.23) -0.12 (0.20) 0.14 (0.16) 0.02 (0.23)
Competitor’s price t=-3. positive 0.14 (0.12) 0.02 (0.08) -0.16** (0.06) -0.03 (0.11) 0.08 (0.22) 0.15 (0.19) 0.11 (0.15) -0.53** (0.23)
Competitor’s price t=-4. positive -0.18 (0.13) -0.26*** (0.09) -0.07 (0.07) -0.10 (0.12) 0.22 (0.20) 0.17 (0.19) 0.03 (0.15) 0.22 (0.23)
Competitor’s price t=--1. negative -0.01 (0.13) 0.22 (0.21) 0.09 (0.07) 0.09 (0.13) 0.22 (0.20) 0.13 (0.23) -0.06 (0.25) 0.33 (0.29)
Competitor’s price t=--1. negative 0.06 (0.14) -0.27 (0.21) 0.07 (0.06) -0.01 (0.12) 0.11 (0.23) 0.18 (0.24) -0.07 (0.25) -0.01 (0.29)
Competitor’s price t=--3. negative 0.01 (0.14) 0.00 (0.21) 0.07 (0.06) 0.09 (0.11) -0.12 (0.24) 0.56 (0.23) 0.09 (0.26) 0.36 (0.31)
Competitor’s price t=--4. negative -0.17 (0.13) -0.19 (0.22) -0.01 (0.05) -0.09 (0.12) 0.29 (0.23) 0.25** (0.24) -0.08 (0.28) 0.50 (0.31)
Electricity price t=0 0.09 (0.06) 0.12*** (0.05) -0.01 (0.03) -0.51* (0.28) 0.04 (0.12) 0.12 (0.08) -0.06 (0.06) 0.04 (0.09)
Electricity price t=-1 0.06 (0.06) 0.01 (0.05) -0.05 (0.03) 0.21 (0.28) 0.39*** (0.12) 0.10 (0.08) 0.10* (0.06) -0.01 (0.09)
Electricity price t=-2 -0.04 (0.06) -0.07 (0.05) -0.03 (0.03) 0.51* (0.30) -0.05 (0.13) 0.08 (0.09) 0.05 (0.06) 0.01 (0.09)
Electricity price t=-3 -0.12* (0.06) -0.08 (0.05) -0.01 (0.03) 0.20 (0.29) -0.06 (0.12) 0.02 (0.09) 0.03 (0.07) 0.08 (0.09)
Electricity price t=-4 -0.06 (0.06) 0.06 (0.05) -0.01 (0.03) -0.38 (0.30) -0.13 (0.12) -0.05 (0.09) -0.04 (0.06) -0.11 (0.09)
25
Code GP 2002 211 212 222 232 241 244 245 246
Gas price t=0 0.05* (0.03) -0.04* (0.02) 0.00 (0.01) 0.00 (0.12) 0.06 (0.05) 0.04 (0.03) -0.01 (0.02) 0.08** (0.04)
Gas price t=-1 -0.05** (0.03) -0.03 (0.02) 0.01 (0.01) 0.10 (0.12) -0.03 (0.05) -0.05 (0.03) -0.02 (0.02) -0.01 (0.04)
Gas price t=-2 -0.02 (0.02) 0.01 (0.02) 0.01 (0.01) -0.19 (0.11) -0.02 (0.05) -0.01 (0.03) 0.03 (0.02) 0.04 (0.03)
Gas price t=-3 -0.01 (0.03) 0.04** (0.02) 0.00 (0.01) 0.19 (0.12) 0.03 (0.05) -0.02 (0.03) 0.00 (0.02) -0.07* (0.04)
Gas price t=-4 0.02 (0.02) -0.01 (0.02) -0.01 (0.01) -0.05 (0.11) -0.01 (0.04) 0.07** (0.03) 0.02 (0.02) 0.03 (0.04)
Oil price t=-0 0.00 (0.01) 0.00 (0.00) 0.00 (0.00) 0.43*** (0.03) 0.02* (0.01) 0.00 (0.01) 0.01 (0.01) 0.00 (0.01)
Oil price t=-1 -0.01 (0.01) 0.00 (0.00) 0.00 (0.00) 0.01 (0.08) 0.02** (0.01) -0.02** (0.01) -0.01 (0.01) 0.01 (0.01)
Oil price t=-2 0.01 (0.01) 0.01*** (0.00) 0.00 (0.00) 0.14* (0.08) 0.01 (0.01) 0.00 (0.01) 0.00 (0.01) -0.01 (0.01)
Oil price t=-3 0.01 (0.01) 0.01** (0.01) -0.01* (0.00) -0.05 (0.07) 0.01 (0.01) -0.01 (0.01) 0.00 (0.01) -0.01 (0.01)
Oil price t=-4 0.00 (0.01) 0.01** (0.01) 0.00 (0.00) 0.14* (0.08) 0.01 (0.01) 0.00 (0.01) 0.00 (0.01) 0.01 (0.01)
Coal price t=0 0.00 (0.01) -0.01 (0.01) -0.01 (0.01) 0.00 (0.07) 0.01 (0.03) 0.00 (0.02) 0.00 (0.01) 0.00 (0.02)
Coal price t=-1 0.00 (0.02) -0.01 (0.01) 0.00 (0.01) -0.08 (0.08) 0.05 (0.03) 0.04 (0.02) 0.00 (0.02) -0.02 (0.02)
Coal price t=-2 0.00 (0.02) -0.02 (0.01) 0.00 (0.01) -0.10 (0.08) -0.02 (0.03) 0.02 (0.02) 0.01 (0.01) 0.01 (0.02)
Coal price t=-3 -0.01 (0.02) 0.00 (0.01) 0.00 (0.01) 0.14* (0.08) 0.05 (0.03) -0.04* (0.02) 0.00 (0.01) 0.04* (0.02)
Coal price t=-4 0.02 (0.02) 0.01 (0.01) 0.00 (0.01) 0.03 (0.07) -0.04 (0.03) 0.00 (0.02) 0.00 (0.01) 0.00 (0.02)
EU ETS dummy 0.00 (0.00) 0.00 (0.01) 0.00 (0.00) 0.00 (0.12) 0.00 (0.00) 0.00** (0.00) 0.00 (0.00) 0.00** (0.00)
Observation
R2
F-Test
AIC
BIC
103
0.64
3.69***
-8.06
-7.19
103
0.52
2.30***
-8.66
-7.80
103
0.40
1.40
-9.48
-8.61
103
0.87
13.67***
-4.95
-4.08
103
0.71
5.06***
-6.78
-5.91
103
0.38
1.26
-7.53
-6.66
103
0.26
0.77
-8.18
-7.83
103
0.47
1.86**
-7.31
-6.44
Note: *. ** and *** show significance at the 10%-. 5%-. and 1%-level. respectively. Standard errors are indicated in brackets.
26
Table 5: Regression results in the energy-intensive EU ETS sectors: GP code 25 -27 (4 lag)
Dependent variable:
Sectors Rubber and Plastic Non-metallic mineral products Basic Metals
Code GP 2002 251 252 261 262 263 265 268 271 274
Constant 0.00 (0.00) 0.00* (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.01)
Own price-1 -0.07 (0.12) 0.28** (0.13) 0.01 (0.13) 0.06 (0.12) -0.04 (0.12) 0.00 (0.12) -0.33*** (0.12) -0.22 (0.17) -0.87*** (0.20)
Own price-2 -0.20 (0.12) 0.16 (0.13) 0.19 (0.12) 0.10 (0.11) -0.24** (0.12 ) 0.17 (0.12) -0.03 (0.14) -0.21 (0.18) -0.08 (0.24)
Own price-3 -0.08 (0.12) -0.26* (0.13) 0.09 (0.13) 0.00 (0.11) 0.19 (0.12) -0.03 (0.12) -0.02 (0.14) 0.21 (0.18) 0.10 (0.24)
Own price-4 0.01 (0.13) 0.04 (0.14) 0.10 (0.13) 0.08 (0.12) 0.11 (0.12) 0.12 (0.12) -0.01 (0.14) -0.13 (0.19) -0.07 (0.22)
Competitor’s price t=-1. positive 0.17 (0.17) 0.04 (0.19) 0.25 (0.21) 0.06 (0.15) 0.15 (0.23) 0.00 (0.15) 0.27 (0.48) 0.98*** (0.23) 1.14*** (0.24)
Competitor’s price t=-2. positive 0.33* (0.17) -0.20 (0.22) 0.02 (0.19 ) 0.10 (0.15) 0.02 (0.22) -0.06 (0.15) -0.67 (0.49) 0.22 (0.27) -0.07 (0.27)
Competitor’s price t=-3. positive 0.06 (0.17) -0.04 (0.22) -0.06 (0.20) -0.20 (0.16) -0.05 (0.23) 0.06 (0.15) 0.31 (0.48) 0.31 (0.27) 0.42 (0.27)
Competitor’s price t=-4. positive 0.10 (0.17) -0.30 (0.22) -0.10 (0.20) 0.00 (0.16) -0.03 (0.23) -0.06 (0.15) -0.19 (0.45) -0.17 (0.27) -0.20 (0.24)
Competitor’s price t=--1. negative 0.29 (0.28) 0.25 (0.25) 0.07 (0.48) -0.03 (0.22) 0.09 (0.38) 0.07 (0.07) 0.32 (0.87) 0.57* (0.31) 0.92*** (0.23)
Competitor’s price t=--1. negative 0.00 (0.27) 0.12 (0.25) -0.49 (0.49) -0.20 (0.22) -0.33 (0.38) 0.02 (0.08) 0.02 (0.82) -0.22 (0.38) 0.10 (0.25)
Competitor’s price t=--3. negative 0.00 (0.22) 0.13 (0.25) 0.42 (0.46) 0.23 (0.21) -0.12 (0.35) 0.22*** (0.08) 1.29 (0.84) -0.25 (0.37) -0.33 (0.25)
Competitor’s price t=--4. negative -0.12 (0.22) 0.08 (0.25) 0.84* (0.48) -0.06 (0.21) -0.35 (0.41) -0.16** (0.08) -1.26 (0.90) 0.01 (0.34) 0.33 (0.24)
Electricity price t=0 0.12** (0.06) 0.03 (0.04) 0.03 (0.07) 0.14** (0.06) 0.14 (0.09) 0.55*** (0.11) 0.08 (0.17) 0.01 (0.24) -0.04 (0.32)
Electricity price t=-1 0.03 (0.06) 0.01 (0.04) -0.04 (0.08) -0.07 (0.06) -0.10 (0.09) 0.01 (0.13) -0.13 (0.17) 0.00 (0.25) 0.49 (0.31)
Electricity price t=-2 0.01 (0.06) -0.02 (0.04) -0.01 (0.08) 0.01 (0.07) 0.02 (0.08) -0.07 (0.13) -0.18 (0.18) 0.13 (0.26) -0.22 (0.33)
Electricity price t=-3 0.06 (0.06) -0.01 (0.04) -0.20** (0.08) 0.18*** (0.06) -0.10 (0.09) -0.09 (0.12) 0.18 (0.18) 0.15 (0.25) 0.19 (0.31)
Electricity price t=-4 0.08 (0.07) -0.05 (0.04) 0.04 (0.08) -0.16** (0.07) 0.03 (0.09) -0.10 (0.12) 0.05 (0.18) -0.25 (0.24) 0.23 (0.32)
Gas price t=0 -0.02 (0.02) -0.01 (0.02) -0.02 (0.03) -0.09*** (0.02) -0.01 (0.03) -0.09** (0.04) 0.03 (0.07) 0.14 (0.10) 0.04 (0.13)
27
Gas price t=-1 -0.01 (0.02) 0.02 (0.02) 0.01 (0.03) 0.03 (0.03) 0.01 (0.03) -0.06 (0.05) -0.02 (0.08) -0.02 (0.10) -0.13 (0.14)
Gas price t=-2 0.00 (0.02) 0.00 (0.02) -0.03 (0.03) 0.05 (0.02) -0.01 (0.03) 0.01 (0.04) 0.01 (0.07) -0.03 (0.09) 0.01 (0.12)
Gas price t=-3 0.04* (0.02) 0.00 (0.02) 0.05 (0.03) 0.00 (0.03) 0.03 (0.03) 0.06 (0.04) -0.02 (0.08) -0.10 (0.10) 0.07 (0.13)
Gas price t=-4 -0.03 (0.02) 0.01 (0.02) -0.02 (0.03) 0.00 (0.02) -0.03 (0.03) 0.03 (0.04) 0.03 (0.07) 0.06 (0.09) 0.05 (0.12)
Oil price t=-0 0.00 (0.01) 0.00 (0.00) -0.01 (0.01) 0.00 (0.01) -0.01 (0.01) 0.01 (0.01) 0.04*** (0.02) 0.05** (0.02) 0.06* (0.03)
Oil price t=-1 0.00 (0.01) 0.00 (0.00) -0.01 (0.01) 0.01 (0.01) 0.01 (0.01) 0.00 (0.01) 0.01 (0.02) -0.01 (0.03) -0.04 (0.03)
Oil price t=-2 0.00 (0.01) 0.01* (0.00) 0.00 (0.01) 0.01* (0.01) 0.00 (0.01) 0.01 (0.01) 0.00 (0.02) -0.01 (0.02) -0.01 (0.03)
Oil price t=-3 0.01 (0.01) 0.01 (0.00) 0.00 (0.01) 0.01 (0.01) -0.01 (0.01) -0.01 (0.01) 0.00 (0.02) -0.02 (0.03) 0.01 (0.04)
Oil price t=-4 0.01 (0.01) 0.00 (0.00) 0.00 (0.01) 0.01 (0.01) -0.02* (0.01) 0.00 (0.01) 0.01 (0.02) -0.01 (0.03) 0.07* (0.04)
Coal price t=0 0.00 (0.01) 0.02 (0.01) 0.06*** (0.02) 0.01 (0.01) 0.01 (0.02) 0.04* (0.02) -0.02 (0.04) 0.05 (0.06) 0.08 (0.07)
Coal price t=-1 0.01 (0.01) -0.01 (0.01) -0.05** (0.02) -0.01 (0.02) -0.04** (0.02) -0.03 (0.03) -0.02 (0.04) -0.07 (0.06) -0.06 (0.09)
Coal price t=-2 -0.01 (0.01) 0.00 (0.01) 0.02 (0.02) -0.01 (0.02) 0.03 (0.02) 0.01 (0.03) -0.01 (0.04) -0.02 (0.06) 0.11 (0.08)
Coal price t=-3 0.00 (0.02) 0.00 (0.01) -0.01 (0.02) 0.02 (0.02) 0.00 (0.02) 0.05* (0.03) 0.00 (0.05) 0.12* (0.07) -0.20** (0.09)
Coal price t=--4 0.00 (0.01) 0.00 (0.01 ) -0.02 (0.02) -0.01 (0.02) 0.01 (0.02) 0.02 (0.03) 0.03 (0.04) -0.10* (0.06) -0.03 (0.09)
EU ETS dummy -0.02 (0.02) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00) 0.00** (0.00) 0.00* (0.00) 0.00 (0.00) 0.00 (0.00) 0.00 (0.00)
Observation
R2
F-Test
AIC
BIC
103
0.28
0.82
-8.16
-7.29
103
0.45
1.73**
-8.79
-7.92
103
0.40
1.42
-7.63
-6.77
103
0.39
1.34
-8.05
-7.18
103
0.41
1.43
-7.51
-6.64
103
0.58
2.92***
-6.91
-6.04
103
0.39
1.31
-6.00
-5.13
103
0.63
3.57***
-5.31
-4.44
103
0.59
2.97***
-4.71
-3.84
Note: *. ** and *** show significance at the 10%-. 5%-. and 1%-level. respectively. Standard errors are indicated in brackets.